.
I've been thinking a fair bit about settlers recently. As I see it, a new city has three costs: the cost of the settler, the cost of its defense, and the food cost of regenerating the population. Usually we can lump the food and hammer costs together since you can switch tiles and most tiles produce 2 of either. So it's about 20 + 10 + 10 = 40 units if you're in Republic, or 20 + 10 + 30 = 70 units if you're not.
Mainly what prompted this was your comment in one of your articles about how rushing a library is a mistake. If you're pre-Republic, the correct comparison isn't 1 library = 2 settlers, it's more like 1 library = 0.6 setters. Still, I've come to see that it's not a good idea until you've got a city that is using all its food tiles, all its hammer tiles, and is still producing 6+ science with the extra population. Or as a side effect of tiles that produce food (i.e. Japan, Egypt, Spain with whales).
- Gus
Interesting thoughts on this, but it's not very accurate. It's not just about comparing the straight up hammers and gold, and then the total tech output, it's a lot more than that. But, either way, more settlers will beat a library pretty much every time, even with civs that have growth from the sea in the early game.
40 hammers will take too long in the early game to hammer out in a city (other than in Athens with the Greeks, which is one exception to the rule of not building an early library). So, you're spending 8-10 turns building a library in a city if you put it all on hammers, but you are not growing in this city. It simply takes too long to hammer one out, so this usually means your spending somewhere around 80 gold to rush the library if you want it in a timely fashion.
Also, you have to defend this city, which you left out of the equation comparing the two possiblities. A flip, a spy, or someone taking your city will make the damage even worse.
The most important factor is the fact that you have to expand, and the best time to do the bulk of your expansion is in the early phase of the game. The more cities you get out pre-republic, the more you'll be able to easily expand in Republic. 3 cities is about the miniumum you want to have before republic, with 5 being about right for me. If I can get more, great.
But, with 20 hammers, or 40 gold in the ancient era, plus banking 8 food, you can have a city at 2 pop and another settled at 2 or 3 pop, depending on if you settle it still in the ancient era. So, you either have 4 or 5 total workers compared to probably 3 workers in the city with the library. Spending another few turns on food and then rushing another settler, you'll have 5-8 workers, or even more, dpending on how quickly you get your 100 gold. 5 cities, all 2-3 pop is superior to 2-3 cities with one having a library in the BC years, especially in the ancient era. Having more small cities makes it way easier to outexpand the other guy.
The early techs don't mean much, the 20 beaker techs don't give 1st to research bonuses. Masonary is a 30 beaker tech that will give you a free wall, but it's usually in your cap, and doesn't really do much for you, as walls are rarely useful. The defensive boost is ok, but not great.
Techs like irrigation, currency, construction, code of laws, all give nice bonuses, but it's not make or break if you don't get them. Plus, with more small cities vs. 1-2 cities with libraries, I usually still outech the other guys to the early bonuses.
More cities means easier defense, no question. Sometimes you can play stupidly and spread yourself too thin, but this is bad play and decision making. You should be able to stop attackers more easily than the other guy with less cities, because you have more workers.
The game has a flow, and by putting unnecessary priority on an early library, you'll stagnate compared to the other guy who has focused on expanding their empire, and can be way more flexible. You really aren't sacrifcing much to build more settlers, and with each new city settled, the more possiblities are open to you. Tech is only a part of the game, and it's easy for people to put way too much importance on teching, because of all the great bonuses you get, and the abilities it affords in advancing your civilization.
But, there are other equally important things to do, especially in the early game, namely claiming more of the map, taking cities, getting gold, exploring, etc. You can turn production, gold, growth, culture, or units into more power. A library only gets you more beakers, and some overflow gold.
This has been discussed many times, and no experienced top player will recommend rushing the early library. I do every now and then, but it's when I'm lazy and I have more than 200 gold in the Ancient, or more than 250 gold and want to still be able to receive the free market bonus from currnecy (which you will not get if you get Currency from 250 gold milestone).
I see this all the time online, especially Japanese players. They will have 2-4 cities, and they will all have libaries. They might jump to an early tech lead, but often we are equal in tech. But, I start pulling away, and then just leave them in the dust. These aer some of the easiest games.
It is a "rule of thumb", sometimes there are exceptions to this, but it's pretty rare. I recommend playing the first 40 turns over and over again (4000BC-0AD), and try to get 10 cities settled in the first 40 turns. Then compare how you did in that game compared to how you do when you don't expand that much and place more emphasis on buildings.